Propensity score weighting: an application to an Early Head Start dental study.
Propensity score weighting: an application to an Early Head Start dental study.
复制标题
DOI：
10.1111/jphd.12106
复制
发表时间：
2016
影响因子：
2.3
通讯作者：
Rozier RG
中科院分区：
文献类型：
--
作者：
Burgette JM;Preisser JS;Rozier RG
干预研究中的非随机组分配可能会导致干预前协变量的不平衡和有偏见的效果估计。我们使用倾向分数估计来解释具有丰富的预处理信息的早期启动(EHS)数据集中的这种失衡。我们比较了使用标准Logistic回归模型(LRM)和广义助推模型(GBM)的倾向评分结果。我们使用47个社会人口学特征和父母访谈获得的EHS登记标准来估计倾向分数，这些标准来自全州范围内637个EHS和930个与Medicaid匹配的对照儿童的样本。LRM和GBM被用来估计与EHS登记相关的倾向分数。这两种方法的性能都是通过测量治疗前协变量分布在治疗前和对照对象之间的平衡以及通过有效样本量衡量的倾向性得分权重的稳定性来评估的。所有变量在EHS组和非EHS组的分布通过使用LRM和GBM计算的倾向分数权重来平衡。与LRM相比，GBM在治疗和倾向分数加权控制分布之间取得了更好的平衡。对照组的有效样本量从930人降至507人(GBM组)和335人(LRM组)。虽然GBM和LRM的倾向性得分都有效地平衡了干预前观察到的协变量，但GBM导致了比LRM更好的协变量平衡。与LRM相比，GBM还导致控制组的有效样本量更大。在这项EHS干预研究中，由于测量的干预前协变量分布不平衡，使用GBM进行倾向性分数加权是减少混淆的有效统计方法。
Non-randomized group assignment in intervention studies can lead to imbalances in pre-intervention covariates and biased effect estimates. We use propensity score estimation to account for such imbalances in an Early Head Start (EHS) dataset with rich pretreatment information. We compare propensity score results using standard logistic regression models (LRM) versus generalized boosted models (GBM). We estimated propensity scores using 47 socio-demographic characteristics and EHS enrollment criteria obtained by parent interviews from a state-wide sample of 637 EHS and 930 Medicaid-matched control children. LRM and GBM were used to estimate propensity scores related to EHS enrollment. Performance of both approaches was evaluated via measures of balance of pre-treatment covariate distributions between treated and control subjects; and stability of propensity score weights measured by the effective sample size. Distributions of all variables were balanced for EHS and non-EHS groups using propensity score weights calculated with LRM and GBM. Compared to LRM, GBM resulted in better balance between treated and propensity score weighted control distributions. The effective sample size of the controls decreased from 930 subjects to 507 with GBM and to 335 with LRM. Although propensity scores derived from GBM and LRM both effectively balanced observed pre-intervention covariates, GBM resulted in better covariate balance compared to LRM. GBM also resulted in a larger effective sample size of the control group compared to LRM. Propensity score weighting using GBM is an effective statistical method to reduce confounding due to imbalanced distributions of measured pre-intervention covariates in this EHS intervention study.